sign in sign up
<<
Home , Condensin, Nuclear lamina, Nuclear pore

INVESTIGATION

Condensins Exert Force on - Tethers to Mediate Nucleoplasmic Reticulum Formation in

Julianna Bozler,* Huy Q. Nguyen,* Gregory C. Rogers,† and Giovanni Bosco*,1 *Geisel School of Medicine at Dartmouth, Hanover, New Hampshire 03755, and †Department of Cellular and Molecular Medicine, University of Arizona Cancer Center, University of Arizona, Tucson, Arizona 85724

ABSTRACT Although the nuclear envelope is known primarily for its role as a boundary between the KEYWORDS nucleus and in , it plays a vital and dynamic role in many cellular processes. Studies of nuclear nuclear structure have revealed tissue-specific changes in nuclear envelope architecture, suggesting that its architecture three-dimensional structure contributes to its functionality. Despite the importance of the nuclear envelope, chromatin force the factors that regulate and maintain nuclear envelope shape remain largely unexplored. The nuclear nucleus envelope makes extensive and dynamic interactions with the underlying chromatin. Given this inexorable chromatin link between chromatin and the nuclear envelope, it is possible that local and global chromatin organization compaction reciprocally impact nuclear envelope form and function. In this study, we use Drosophila salivary glands to nuclear envelope show that the three-dimensional structure of the nuclear envelope can be altered with II- mediated chromatin condensation. Both naturally occurring and engineered chromatin-envelope interac- tions are sufficient to allow chromatin compaction forces to drive distortions of the nuclear envelope. Weakening of the further enhanced envelope remodeling, suggesting that envelope struc- ture is capable of counterbalancing chromatin compaction forces. Our experiments reveal that the nucle- oplasmic reticulum is born of the nuclear envelope and remains dynamic in that they can be reabsorbed into the nuclear envelope. We propose a model where inner nuclear envelope-chromatin tethers allow inter- phase movements to change nuclear envelope morphology. Therefore, chromatin compaction may be a normal mechanism that reorganizes nuclear architecture, while under pathological conditions, such as , compaction forces may contribute to defects in nuclear morphology.

During the past several decades we have begun to appreciate the DNA damage repair, and response to mechanosensory inputs (De structural and functional complexity of the nuclear envelope. Although Vosse et al. 2011; Maniotis et al. 1997; Therizols et al. 2006). The primarily known for its role in the partitioning of genomic and tran- well-described physical structures of the double membrane layer and scriptional components away from the cytoplasm, the nuclear envelope nuclear structural filaments, such as nuclear , remain the pre- plays a vital and dynamic role in expression, signal transduction, dominant platforms for interaction between chromatin and the cyto- plasm (Shimi et al. 2012). These two membrane layers are connected Copyright © 2015 Bozler et al. through integral membrane and are punctuated with the doi: 10.1534/g3.114.015685 complex that facilitates bidirectional macromolecule Manuscript received November 12, 2014; accepted for publication December 22, movement between and cytoplasm. The internal struc- 2014; published Early Online December 30, 2014. tural support of the nucleus is provided in large part by the nuclear This is an open-access article distributed under the terms of the Creative Commons Attribution Unported License (http://creativecommons.org/licenses/ lamina, a meshwork of A- and B-type polymers. Studies of by/3.0/), which permits unrestricted use, distribution, and reproduction in any nuclear structure have exposed tissue- specificchangesinthe medium, provided the original work is properly cited. threee-dimentional architecture of the nuclear envelope, revealing in- Supporting information is available online at http://www.g3journal.org/lookup/ tricate patterns of tunneling and branching of the nuclear envelope suppl/doi:10.1534/g3.114.015685/-/DC1 1Corresponding author: Department of Genetics, Geisel School of Medicine within the nuclear space (Dupuy-Coin et al. 1986; Fricker et al. 1997). at Dartmouth, 7400 Remsen, Hanover, NH 03755. E-mail: Giovanni. This higher order arrangement of the nuclear envelope has been given [email protected] the name nucleoplasmic reticulum (NR) (Malhas et al. 2011). The

Volume 5 | March 2015 | 341 importance of these NR structures for chromosome partitioning, nu- Cytoskeletal forces have been studied in the context of nuclear clear transport, , or other cellular processes are not architecture (Maniotis et al. 1997), but the role of chromatin has only well understood. Although we lack a comprehensive understanding of begun to be implicated in the mechanical stabilization of nuclear the functional role of nuclear architecture, it has become clear that the architecture. Recent studies have revealed that decondensation of chro- nuclear envelope and its three-dimensional shape can profoundly matin directly leads to a mechanically destabilized nucleus (Mazumder impact organization and gene expression (De Vosse et al. et al. 2008). Additionally, it has been shown that chromatin acts as 2011; Schneider and Grosschedl 2007). a force-bearing element in the unchallenged nucleus (Dahl et al. 2005). The typical interphase genome is radially organized with hetero- This has led to the supposition that nuclear and cytoplasmic anchors chromatin and transcriptionally silent gene regions predominantly are important contributors to the structural integrity of the nucleus. restricted to the nuclear periphery (Hochstrasser and Sedat 1987; Further, it has been suggested that chromatin-envelope interactions Rajapakse and Groudine 2011). However, exceptions to this trend constitute a mechanical scaffold that defines and preserves nuclear exist. For example, nuclear pore complexes can recruit and tether architecture (Mazumder and Shivashankar 2007; Mazumder et al. ı transcriptionally active (Capelson et al. 2010; Rodr guez- 2008). Mounting evidence suggests that mechanical forces within the Navarro et al. 2004). These observations have led to a model in which nucleus may have an equally profound effect on nuclear structure and association with the nuclear envelope serves as a boundary between yet remain relatively enigmatic (Buster et al. 2013; George et al. 2014). active and inactive chromatin (Bermejo et al. 2011; Blobel 1985). Importantly, dynamic forces exist within the nucleus, not only in Other models propose that chromatin tethers to the nuclear envelope preparation for division, but during interphase as well. The serve to organize into distinct territories (Bauer et al. condensin complexes are recognized as regulators of chromatin 2012; Verdaasdonk et al. 2013). Regardless of the functional role structure and topology during , with condensin II chromatin-envelope interactions may have, it is clear that chromatin maintaining activity during interphase. Shared between condensin I is tethered to the inner nuclear envelope through a variety of DNA- and condensin II is the SMC2/4 heterodimer, endowing the complex and protein-protein interactions (Dahl et al. 2005; Gruenbaum with its essential ATPase domains (Kimura and Hirano 1997). Unique et al. 2005; Mazumder et al. 2008). Chromatin interacts with the to condensin II are the chromosome-associating protein subunits nuclear envelope through contacts with the nuclear lamins or the Cap-D3, Cap-H2, and Cap-G2, the regulation of which likely confers nuclear pore complex (Ye et al. 1997). Lamins bind chromatin through fi interphase activity (Buster et al. 2013; Hartl et al. 2008; Hirano et al. interactions with or by directly binding sequence-speci c 1997). During interphase, condensin II participates in the mainte- regions of DNA termed matrix-attachment regions (Zhao et al. nance of and drives homolog unpairing 1996). These chromatin-envelope tethers facilitate gene regulation (Bauer et al. 2012; Hartl et al. 2008). Hyperactivation of condensin as well as genome organization (Dialynas et al. 2010; Therizols et al. II, either by inactivation of the negative-regulator SCFSlimb or by over- 2006; Verdaasdonk et al. 2013). expression of Cap-H2, results in dramatic chromosome condensation, Given the substantial role of the nuclear envelope in genome suggesting that dynamic chromatin movements occur during inter- regulation, it is not surprising that defects in nuclear architecture are phase and that these movements are induced by condensin II activity linked to a variety of diseases. There are 12 known disorders that arise (Bauer et al. 2012; Hartl et al. 2008). Similarly, loss of condensin II from mutations in the A-type lamins and lamin-associated proteins, function leads to lengthening of interphase chromosomes (Bauer et al. such as Emery-Dreifuss muscular dystrophy and Hutchinson-Gilford 2012). Therefore, condensin II could be seen as more broadly regu- syndrome (Goldman et al. 2004; Sullivan et al. 1999). In these lating the mechanical properties of chromatin mediated support of the syndromes, the characteristic deformation of the nuclear envelope results in a loss of peripheral , altered gene expression nuclear envelope. Notably, in Drosophila, interphase condensin II (Goldman et al. 2004), and disorganization of the genome (Goldman hyperactivity induces nuclear envelope deformations (Buster et al. et al. 2004; Shumaker et al. 2006). Interestingly, the accumulation of 2013), and in cancer cell lines, RNA interference (RNAi) de- aberrantly spliced lamins can occur over time in normal tissues, linking pletion of condensin II results in increased nuclear size and distortions similar changes in nuclear structure with the process of ageing (Scaffidi of nuclear shape (George et al. 2014). Hence, it has been proposed that and Misteli 2006). Additionally, numerous cancers are characterized condensin II-mediated mechanical force within the interphase nucleus through their envelope morphology, although in such instances the is capable of pulling nuclear membrane structures into the nucleus direct causes of nuclear defects are less clear (Chow et al. 2012). and contribute to remodeling of nuclear architecture (Wallace and Two major models have surfaced to address the cause of nuclear Bosco 2013). deformations. The first model proposes that the mechanical properties In this study, we use polytene nuclei from Drosophila melanogaster of the lamins themselves, coupled with their asymmetric distribution, salivary glands to investigate whether condensin mediated chromatin are responsible for nuclear ”blebbing” and invagination of the enve- compaction could lead to remodeling of the nuclear envelope and lope (Funkhouser et al. 2013). A second model suggests that the induce NR-like structures. We chose salivary gland cells because in stabilizes nuclear architecture by counterbalancing inter- third instar larvae these cells are postmitotic and never re-enter the nal forces of the nuclear lamina and chromatin, and only when these . Therefore, this eliminates the possibility that envelope reas- stabilizing factors are altered does the nuclear architecture change sembly upon exit from could contribute to alterations in nu- (Dahl et al. 2005; Lammerding et al. 2006; Maniotis et al. 1997; clear envelope structure. Additionally, these nuclei are relatively large, Mazumder et al. 2008). This model allows for dynamic counterbal- making it possible to visualize chromatin-envelope interactions with ancing forces to maintain normal architecture, but which could be standard light microscopy. We observe measurable differences in the regulated to form complex nuclear structures in nondisease tissues NR after chromatin compaction mediated by condensin II. For the (Mammoto and Ingber 2010; Mazumder et al. 2010). Evidence exists first time, the dynamic formation of the NR is observed in live cells, for each model (Rowat et al. 2008); however, a comprehensive model and we find that artificial chromatin tethers are sufficient to pull of nuclear architecture will likely be complex and include several un- nuclear envelope structures into the interior of the nucleus. An in- identified factors. vestigation into the formation of the NR will translate into an

342 | J. Bozler et al. understanding of the dynamic nature of nuclear envelope structure H2-GFP protein, in which the last 23 amino acids of the Cap-H2 protein and reveal forces that regulate nuclear envelope shape and function. have been truncated (UAS-Cap-H2DC23-eGFP.3). This deletes a deg- radation signal recognized by the Slimb E3- (Buster MATERIALS AND METHODS et al. 2013). Body wall dissections were carried out using previously described techniques (Budnik et al. 1990) and stained with anti- Drosophila strains Dm0 Lamin. All Drosophila melanogaster stocks were maintained at 25 on stan- A Nikon A1R SI Confocal microscope was used for all light dard corn meal molasses media, with a 12-hr light/dark cycle. As in microscopy. Image averaging of 4x during image capture was used for previous studies, increased condensin II activity was induced by over- all images unless otherwise specified. The Nikon software Nikon expressing a single subunit, Cap-H2 (Bauer et al. 2012; Hartl et al. 2008). Elements (V 2.0) was used for three-dimentional projections and for Two different methods of Cap-H2 overexpression are described here, three-dimentional analysis of the nuclear envelope. both of which use the GAL4/UAS system. A heat shock2inducible system was used to allow for precise temporal control of Cap-H2 Quantification and characterization of NR in overexpression. In this fly line, the Hsp70-GAL4 is able to drive salivary glands expression of endogenous Cap-H2 due to a UAS insertion upstream To quantify the observed NR phenotype, immunofluorescence was of the Cap-H2 gene (Bauer et al. 2012; Hartl et al. 2008). This performed on Cap-H2 overexpressing larvae and compared to control. . . Cap-H2 overexpression line, referred to as the HS GAL4 UAS Vials of third instar larvae of the Cap-H2 overexpression line HS . Ã Cap-H2 line (w ;HS83-GFP-LacI, LacO (60F);Hsp70-GAL4, EY09979), GAL4 UAS . Cap-H2, and the control line HS . GAL4, underwent is the primary method of Cap-H2 overexpression used in this study. heat shock for 1.5 hr at 37. Salivary glands were dissected in PBS Ã The corresponding control line, lacking the EY09979 (w ;HS83- after a 1.5-hour recovery at room temperature. Primary antibody to . GFP-LacI, LacO (60F);Hsp70-GAL4), is denoted as the HS Lamin as described previously was used for visualization of the nu- GAL4 line. clear envelope and NR. Five glands were imaged for each of three fi To con rm that the observations made are not dependent on heat replicate. Z-stacks were taken for 10 nuclei from each salivary gland, fi shock, and to lend further evidence that the phenotype is not speci c with step size of 0.5 microns. An NR event was tallied when a distinct to the particular overexpression transgene, the salivary gland driver three-dimentional invagination of the envelope into the nuclear space . fi 43B GAL4 was used to express Cap-H2 in a tissue-speci c manor. was identified, with a minimum size cut off of 1 mm; superficial fl This driver line was crossed to transgenic ies containing a UAS wrinkles and folds of the nuclear envelope were not considered. The construct expressing a Cap-H2-GFP protein (UAS-Cap-H2-eGFP.1) number of NR events per nucleus for each gland was determined, and fl (Buster et al. 2013). A detailed list of all y stocks used can be found aWelch’s t-test was conducted on the means of the two groups to in Supporting Information, Table S1. assess statistical significance. Five additional glands from non-heat shock larvae of each line were imaged in parallel for one replicate. Immunofluorescence and microscopy The diameter of NR was classified as the greatest two-dimensional All samples prepared for immunofluorescence were fixed in 4% distance of the NR cross-section. Measurements were taken by Nikon methanol-free formaldehyde in phosphate-buffered saline (PBS; Elements for every NR event observed in the HS . GAL4 and HS . 0.001% triton X) for 5min. Samples were washed in PBS (0.1% GAL4 UAS . Cap-H2 lines. Welch’s T-test was performed to test triton X), and blocked with 2% normal goat serum for 2 hr. Primary significance. antibodies were incubated overnight at 4 in 2% normal goat serum. For A comparison of NR quantification techniques was performed to visualization of the nuclear envelope, mouse anti-Dm0 Lamin was used assess the reliability of NR detection for both the anti-Nup and anti- (Adl 84.12 c; Hybridoma Bank) at a concentration of 1:200, and mouse Lamin antibodies. Salivary glands were collected from larvae of the antibody to the nuclear pore complex was used (mab414; Abcam) at same heat shock treatment and stained with either anti-Nup or anti- a concentration of 1:2500. Secondary antibodies were either fluorescein Lamin. For each group, three glands were imaged, totaling ten nuclei isothiocyanate or Cy3 conjugated (Jackson ImmunoResearch), used at per gland. a concentration of 1:100 and 1:150, respectively, and were incubated To visualize nuclear-localizing GFP in relation to NR, flies with the for 2 hr at room temperature. All salivary gland and cell culture UAS . GFP.nls construct were crossed to the HS . GAL4 UAS . experiments were carried out in this fashion unless otherwise spec- Cap-H2 and HS . GAL4 lines. Third instar larvae were dissected and ified. For experiments which depended on heat shock induction, stained following the described method after a 2-hr heat shock and vials of late third instar larvae experienced heat shock at 37 for 6-hr recovery period. 2 hr, followed by a 2-hr recovery at room temperature; experiments that deviate from this protocol carry specific notation. Detection and quantification of NR in cell culture The membrane marker wheat germ agglutinin (WGA) was used as Kc cells were maintained in Sf900II media (Life Technologies) at an additional membrane stain for select samples, following antibody a constant temperature of 25. To achieve increased condensin II preparation. Alexa Fluor 647 and Alexa Flour 488 conjugated WGA activity in cell culture, Slimb, the negative regulator of Cap-H2 was (Life Technologies) was incubated with a sample for 30 min at room depleted with RNAi (Buster et al. 2013). Control (sk) and gene- temperature at a concentration of 10 mg/mL. All samples were counter specific (Slimb) double-stranded RNA (dsRNA) was generated using stained with DAPI (36 mM), mounted in VECTASHIELD, and stored the following previously described forward and reverse primer at 220 until imaged. pairs (Buster et al. 2013); sk (59CGCTTTTCTGGATTCATCGAC, Muscle nuclei of HS . GAL4 UAS . Cap-H2 and HS . GAL4 59TGAGTAACCTGAGGCTATGG), Slimb (59GGCCGCCACAT lines were compared after a 2-hr heat shock and 6-hr recovery. Experi- GCTGCG, 59CGGTCTTGTTCTCATTGGG). Kc cells, at 50–90% ments using the larval muscle2specific driver C57 GAL4 without heat confluency, were treated in parallel with either control or gene- shock induction also were conducted. This driver line was crossed to specific dsRNA. Every other day, cells were treated with 10 mgof transgenic flies containing a UAS construct expressing a stabilized Cap- dsRNA in 1 mL of culture media. The experiment was terminated on

Volume 5 March 2015 | Chromatin Force Remodels the Nuclear Envelope | 343 the seventh day of treatment, and cells were collected for staining. with antibody to Lamin for identification of the nuclear envelope and Treated cells were allowed to adhere to a concanavlin A2coated NR. Five nuclei were imaged per salivary gland; 15 glands were imaged glass cover slip for 15 min before the media was removed and cover for all experiments, with five glands for each of the three replicates. slip was washed three times with PBS. Cells were then fixed in 4% Quantification and statistical techniques were otherwise identical to the methanol-free formaldehyde and stained for the nuclear pore com- preceding protocol. Raw data is available in supplemental information, plex (anti-Nup) and WGA and counterstained with DAPI. An NR File S5. event was defined as any distinct structure within the nucleus that was positive for both WGA and anti-Nup. Cells from three RNAi Live cell imaging experiments were stained and imaged for each treatment condition, Live cell imaging was conducted on salivary gland nuclei. To view the with a minimum of 75 cells counted for each replicate. Z-stacks were nuclear envelope without fixation, homozygous stocks were made compiled from confocal images with a step size of 0.25 microns. The with a GFP-tagged nuclear pore protein, GFP-Nup107 (Bloomington average number of NR events per nucleus was calculated for each Stock Center stock number 35514), and the EY09979 and heat shock replicate. A Welch’s t-test was performed to assess significance. promoter elements from the HS . GAL4 UAS . Cap-H2 line. Late third instar larvae experienced heat shock at 37 for 30 min in stan- Transmission electron microscopy dard Drosophila media vials. Vials were then removed from the To capture a detailed image of the membrane structure of the NR, transmission electron microscopy was performed on the Cap-H2 overexpression line HS . GAL4 UAS . Cap-H2 and control line HS . GAL4. Vials of third instar larvae from each line underwent heat shock for 1.5 hr at 37. Salivary glands were dissected in PBS after a 2-hr recovery period at room temperature. Salivary glands were fixed overnight in a solution of 2% glutaraldehyde/2% paraformaldehyde. Samples were then prepared for sectioning following a detailed pro- tocol optimized to preserve the ultrastructure of the tissue, which can be found in the supplemental material (File S4). Thin sections were taken and stained with uranyl acetate aqueous solution and lead cit- rate. Imaging was performed on a FEI Tecnai F20ST field emission gun transmission electron microscope.

Chromatin-envelope tether To determine whether tethering of chromatin to the nuclear envelope impacts NR formation, we introduced a LacI-Lamin C fusion protein into our system (a gift from Laurie Wallrath). This fusion protein is able to relocate a LacO array to the nuclear periphery by binding to the LacO sequence and integrating its C terminus into the nuclear lamina (Dialynas et al. 2010). Fly stocks with this tether were crossed to the heat shock-inducible Cap-H2 overexpression line LacO, GFP- LacI; HS . GAL4 UAS . Cap-H2, and the control line LacO, GFP- LacI; HS . GAL4. To account for the frequency of chance associations between the NR and LacO array, an additional cross between LacO, GFP-LacI; HS . GAL4 UAS . Cap-H2 and a control line was tested. Six salivary glands were dissected for each replicate. Nuclear envelope was visualized by immunofluorescence (IF) using nuclear pore antibody, LacO arrays were visualized by green fluorescent protein (GFP) fluorescence of GFP-LacI introduced from the HS . GAL4 UAS . Cap-H2 and HS . GAL4 line. A minimum of 12 nuclei from each gland were imaged, with z-stacks taken at 0.5-mm steps. An association between the LacO array and NR was consid- Figure 1 Changes in nuclear architecture are induced by Cap-H2 ered to be present if the GFP signal was observed at any point along overexpression. Individual salivary gland nuclei with different heat the periphery of the NR event. The proportion of nuclei positive for shock treatments are visualized with DAPI and anti-lamin marking the NR events associating with the LacO array was calculated for each chromatin and nuclear envelope, respectively. Normal nuclear mor- genotype, and x2 analysis conducted to assess statistical significance. phology can be seen in the heat shock inducible Cap-H2 over- expression line in the absence of heat shock treatment (A). Nuclear Weakening of the nuclear envelope deformations can be observed in the heat shock2inducible Cap-H2 Structural integrity of the nuclear envelope was compromised through overexpression line with heath shock treatment (B), where the nuclear lamina has formed structures within the nuclear space. Normal mor- expression of an aberrant lamin protein. Expression of aberrant lamin phology is observed in the HS . GAL4 control line without (C) and protein was achieved with the use of a UAS- line; a transgenic with (D) heat shock treatment. In the particular z-slice shown for the fl y line that expresses the human progerin protein under UAS control heat shock control (D), the is especially prominent in the (gift from David Tree) (Beard et al. 2008). These flies were crossed to center of the nucleus but is not indicative of a global reorganization the HS . GAL4 UAS . Cap-H2 and HS . GAL4 lines; salivary of the chromatin to the nuclear periphery. Scale bar is 10 microns for glands from the progeny were dissected. Salivary glands were stained all panels.

344 | J. Bozler et al. Figure 2 Cap-H2 overexpression leads to remodeling of the nuclear membrane. The nuclear membrane of individual salivary gland nuclei is visualized with wheat germ aggluti- nin (WGA) and antibody to the nuclear pore complex (anti-Nup). Formation of nuclear membrane structures within the nucleus can be seen in Cap-H2 overexpression, induced by heat shock (A). Typical spherical nuclear membrane structure, with absence of intra- nuclear structures, can be seen in GAL4 control with heat shock treatment (B). Quan- tification of these structures shows an average number 1.88 per nucleus in Cap-H2 over- expressing cells and 0.04 per nucleus in cells not overexpressing Cap-H2 (C). This is a sig- nificant increase in frequency with Cap-H2 overexpression, P-value 7.1 e27. Maximum diameter of these structures is not statistically different in Cap-H2 induction compared to those formed in the GAL4 control (D), P-value 0.59. Scale bar is 10 microns for all panels.

incubator and kept at room temperature for an additional 30 min. nuclear envelope, salivary glands from the heat shock2inducible SalivaryglandsweredissectedinPBSandplacedonapoly- Cap-H2 overexpression line (HS83-GFP-LacI, LacO(60F);Hsp70- L-lysine2treated cover slip affixed to the bottom of a stainless-steel GAL4, EY09979), referred to as HS . GAL4 UAS . Cap-H2, was slide with a 10-mm hole. The glands were submerged in PBS and compared with the GAL4 control line (HS83-GFP-LacI, LacO(60F); immediately imaged. Imaging sessions lasted for 45 min; z-stacks Hsp70-GAL4), HS . GAL4. These two lines are genetically identical were taken with 2-mm step size every 3 min for the duration of the except that the HS . GAL4 UAS . Cap-H2 line is capable of over- imaging session. Image averaging of 2· was used during image expressing Cap-H2 whereas HS . GAL4 is not (Hartl et al. 2008). In the capture for all live cell time lapses. Control imaging was conducted HS . GAL4 UAS . Cap-H2 line, nuclear envelope shape is normal in in identical fashion but excluded heat shock. the absence of heat shock induction (Figure 1A). In contrast, staining of the nuclear envelope with anti-Lamin revealed complex three-dimensional RESULTS structures within the nucleus after Cap-H2 expression that we refer to as NR invaginations (Figure 1B and Figure S1). NR structures Condensin-mediated forces remodel the were continuous with or immediately adjacent to the nuclear enve- nuclear envelope lope (Figure S1). An identical heat shock treatment of the HS . Increased condensin II activity leads to defects in nuclear architecture GAL4 line (not capable of overexpressing Cap-H2) rarely resulted in in vivo and in cultured Drosophila cells (Buster et al. 2013; Hartl et al. nuclear envelope defects (Figure 1, C and D and Figure 2C). Immu- 2008). In an effort to systematically characterize these changes in the nostaining with an antibody to the nuclear pore complex revealed that

Figure 3 Cap-H2 stabilization leads to NR formation in cell culture. The nuclear membrane of Kc cell nuclei is visualized with wheat germ agglutinin (WGA) and antibody to the nuclear pore complex (anti-Nup). An increase in NR formation can be observed in cells with Cap- H2 stabilization, achieved through Slimb[RNAi] (A). Nuclei from control RNAi treatment maintain typical spherical configuration (B). Quantifica- tion of these structures shows significant in- crease in frequency with Cap-H2 stabilization (C), P-value 0.0025. Scale bars: 10 microns.

Volume 5 March 2015 | Chromatin Force Remodels the Nuclear Envelope | 345 NR invaginations contained integral membrane proteins and stain positive for the membrane marker WGA (Figure 2). Strikingly, NR structures have a notable absence of DNA as well as histones and nuclear GFP (Figure 1B, Figure 2A, and Figure S2C). Similar results were obtained when an alternative Cap-H2 overexpression transgene, Cap-H2-eGFP, was driven by a tissue-specific43B. GAL4 driver that is not heat shock dependent, indicating that neither heat stress nor the specific overexpression construct are contributing factors in the forma- tion of these nuclear structures (Figure S3C).Takentogether,thesedata fit the description of the NR, where the contents of the intranuclear vesicles are contiguous with the cytoplasm and/or excludes contents normally only found inside the nucleus (Malhas et al. 2011). Nuclei from the HS . GAL4 UAS . Cap-H2 line without heat shock displayed no observable nuclear architecture defects (Figure 1A). Quantification of intra-nuclear structures after heat shock treatment in nuclei labeled with anti-Lamin showed a statistically significant in- crease in the presence of these structures with Cap-H2 overexpression as compared with the GAL4 control line (Figure 2C). The number of events per nucleus increased from 0.04 in the HS . GAL4 line, to 1.88 in the HS . GAL4 UAS . Cap-H2 line, yielding a p-value of 7.1 e27. We emphasize that, although rare, these intranuclear structures are observed in wild-type nuclei that are not overexpressing Cap-H2 (Figure 2, C and D). Although greatly increased in frequency, the intranuclear structures observed in salivary glands in no other way appear to be different from those observed in control cells with normal levels of condensin activity, for instance, diameters of the structures were not different between treatments (Figure 2D). Two different antibodies were used to visualize the nuclear membrane and NR, anti-Nup, and anti-Lamin. To confirm that NR detection does not vary depending on the method used, a direct comparison between the antibody stains was performed. The nuclear envelope of salivary glands was labeled with WGA and either anti- Nup or anti-Lamin in CapH 2 overexpression (Figure S4, A and B), and control GAL4 line (Figure S4, C and D). Quantification revealed no significant difference between the number of NR events detected with anti-Nup compared with anti-Lamin in the CapH 2 overex- pression line (2.03 and 2.1 NR events per nucleus), nor in the GAL4 control (0.23 and 0.23 NR events per nucleus) (Figure S4E). Examination of other tissues suggests that the formation of NR structures is not restricted to salivary gland nuclei. Heat shock induction of Cap-H2 in the HS . GAL4 UAS . Cap-H2 line induces similar structures in the diploid nuclei of the body wall muscles of third instar larvae (Figure S5, A and B). Nuclear envelope defects are also observed when Cap-H2 is overexpressed using a larval muscle specificdriver (Figure S5,CandD). NR formation also was explored in a nonpolyploid model, specifically cells in culture. One effective method of increasing overall CapH 2 protein levels in cell culture is to reduce levels of Slimb, the ubiquitin ligase that targets CapH 2 for degradation. By reducing Slimb levels through RNAi, CapH 2 is stabilized and drives condensin II-mediated chromatincompaction(Busteret al. 2013). We used this method to evaluate the ability of CapH 2 to drive NR formation in cells. NR events and the nuclear membrane were detected using WGA and anti-Nup staining in control sk[RNAi]-treated cells, indicating that envelope

Figure 4 Transmission electron microscopy (TEM) imaging of nuclear envelope and nucleoplasmic reticulum (NR) in salivary gland nuclei. nuclear periphery of the GAL4 control (A). Cap-H2 overexpressing TEM of salivary glands reveal the ultrastructure of the nuclear envelope. nuclei have additional structures within the nucleus, termed NR (B). Labeled in the panels are the following features: N (nucleus), C Greater resolution images of the Cap-H2 overexpression nucleus, (C) (cytoplasm), Ch (chromatin), NE (nuclear envelope), and NR (nucleo- and (D), reveal that the NR has an intact double membrane layer. Scale plasmic reticulum). The nuclear envelope can be seen lining the bars: (A) 2 microns, (B) 2 microns, (C) 500 nm, (D) 500 nm.

346 | J. Bozler et al. remodeling is a normal occurrence inthesecells(Figure3,AandB). (Malhas et al. 2011). Type I invaginations contain only the nuclear However, in cells treated with Slimb[RNAi], where CapH 2 protein is lamina and inner membrane layer, requiring a delamination of the stabilized, NR events were increased by more than twofold (Figure 3, A envelope’s two membranes. Alternatively, type II invaginations consist and B). Notably, cells displayed a less-severe response to CapH 2 activity of the complete double membrane layer. Our observation that struc- than salivary gland nuclei; however, this is not entirely unexpected, as the tures induced by condensin activation contain nuclear pore proteins surface area, content, and filament strength of a nucleus will vary as well as label positive for the WGA lectin suggested that they are between tissues; these factors will likely modulate the intrinsic properties double membrane structures (Figure 1 and Figure 2). To further of a cell’s nuclear envelope. Additionally, the two experiments relied on validate whether these structures are NR invaginations and to identify different methods for hyper-activation of CapH 2, one through over- the class of NR present in the salivary gland, we examined NR mem- expression, and the other through a stabilization of endogenous protein. brane structure. Transmission electron microscopy of a sectioned sal- These factors may lead to different amounts of CapH 2 activity and ivary gland was performed to view detailed lipid structures of the therefore result in varied severity of the phenotype. Nonetheless, a signif- nuclear envelope. Control nuclei contained densely stained chromatin icant increase of NR formation was detected between sk[RNAi] and with characteristic banding (Figure 4A). In Slimb[RNAi] treatments, 0.12 compared with 0.29 NR events per nu- contrast, NR structures were readily identified in Cap-H2 overexpress- cleus, respectively (Figure 3C), suggesting that nuclear envelope remod- ing salivary gland nuclei (Figure 4B). Close examination of the NR eling through chromatin compaction is not restricted to highly membrane revealed two distinct lipid bilayers, indicating that the polyploidy tissues. double membrane layer was intact (Figure 4, C and D). Furthermore, densely-stained chromatin was never observed inside NR structures. NR events are type II invaginations We conclude that these structures conform to the previously described The structures of the nuclear envelope that constitute the NR have definition of NR structures, specifically type-II NR events (Malhas been separated into two classes; type I, and type II invaginations et al. 2011), and are consistent with DNA, and nuclear-GFP

Figure 5 Chromatin associates with nucleo- plasmic reticulum (NR) membrane. Transmis- sion electron microscopy (TEM) of salivary glands reveal the ultrastructure of the nucleus in Cap-H2 overexpression tissues. Labeled in the panels are the following features: N (nucleus), C (cytoplasm), Ch (chromatin), NE (nuclear envelope), and NR (nucleoplasmic reticulum). Individual NR events are shown (A2C), with increased magnification on the right hand panels. In each instance, chromatin can be observed abutting the membrane of the NR. Scale bars are 500 nm in all panels.

Volume 5 March 2015 | Chromatin Force Remodels the Nuclear Envelope | 347 signals that are excluded from the interior of these membrane con- in the absence of the tether. Three genotypes were assayed for asso- taining structures (Figure 1, Figure 2, and Figure S2). ciation of the LacO array with NR events: Cap-H2 overexpression (LacO, GFP-LacI; HS . GAL4, UAS . Cap-H2) crossed to flies NR invaginations associate with chromatin-envelope with the tether construct, Cap-H2 overexpression (LacO, GFP-LacI; tethers HS . GAL4, UAS . Cap-H2) crossed to flies without the tether Our model predicts that physical interactions or linkages between the construct, and GAL4 control (LacO, GFP-LacI; HS . GAL4) crossed chromatin and nuclear envelope are responsible for the alterations in to flies with the tether construct. In all genotypes, the same LacO the nuclear envelope following chromatin remodeling. Indeed, with arrays were present and visualized by a GFP-LacI fusion protein. As very few exceptions, all type-II NR structures visualized by trans- the GFP-LacI fusion protein is used for detection in place of LacI- mission electron microscopy associated with chromatin on the nuclear- antibody staining, only concentrated GFP-LacI was detectible, and proximal membrane layer (Figure 5). We emphasize that these type-II therefore presented as a single spot at the LacO array (Buster et al. NR structures have clearly visible double membranes, consistent 2013; Hartl et al. 2008). Coexpression of GFP-LacI with LacI-LaminC with the hypothesis that chromatin tethers in the nuclear envelope did not interfere with formation of the tether, since the LacO array allow NR structures to be pulled into the nuclear interior. To test localized to the nuclear envelope. Frequency of NR events and the whether such attachments mediate NR events, we introduced an proportion of nuclei with the LacO array associating with an NR ectopic chromatin-lamin tether, as previously described (Dialynas invagination were measured for each genotype (Figure 6). A statis- et al. 2010). A LacI-LaminC fusion protein was expressed in a genetic tically significant increase in LacO association with NR was observed background where a LacO array and GFP-LacI also were present, for the HS . GAL4 UAS . Cap-H2 line with tether, occurring in thus introducing a chromatin tether into our condensin overexpres- 33% of nuclei compared with 1.6% of nuclei in HS . GAL4 UAS . sion system. If such a linkage facilitates the formation of NR invagi- Cap-H2 without tether and 0% of nuclei in HS . GAL4 with tether. nations, then we predict that the LacI-LaminC tethered DNA would Although the frequency of LacO-associated NR events was dramat- associate with NR at greater frequency than the same region of DNA ically greater in the Cap-H2 overexpression line with a LacI-LaminC

Figure 6 Chromatin-envelope tethers associ- ate with nucleoplasmic reticulum (NR). Sali- vary gland nuclei were examined for effects of ectopically expressed chromatin-envelope tether in Cap-H2 overexpression and control. The nuclear envelope is visualized with anti- body to the nuclear pore complex (anti-Nup) and the chromatin-envelope tether tracked with green fluorescent protein (GFP)-LacI. In the heat shock control nucleus (A), the LacO array can be seen localizing to the nuclear periphery. The Cap-H2 overexpression nu- cleus has alternate localization, with the LacO array in the nuclear interior and associating with NR event (B). NR event can be observed associating with LacO array in greater detail in a greater magnification view (C). Total number of NR events was quantified for each cross (D). Both Cap-H2 overexpression crosses had increased NR events compared to the GAL4 control line, p-value ,2.2e216. Yet, there was no statistical difference between the Cap-H2 overexpression nuclei with and without chromatin-envelope tether, P-value 0.252. The frequency of association of the LacO array to NR was elevated in the Cap- H2 overexpression with chromatin-envelope tether (E), P-value 0.0031. Scale bars are 10 microns in all panels.

348 | J. Bozler et al. tether, there was no difference in the frequency of total NR events et al. 2013; Maniotis et al. 1997; Rowat et al. 2008). The mechanical between lines with or without the LacI-LaminC tether (Figure 6). stress of these forces is distributed through the nuclear envelope, This argues that the LacI-LaminC itself does not affect the frequency with the nuclear lamina adding further structural stability and me- of NR events, and strongly suggests that LacI-LaminC tether-associated diating the interaction of these forces. We postulated that if mechan- NR events were not random. Thus, we conclude that condensin II ical forces provided by chromatin compaction contributes to NR compaction forces act on chromatin-envelope tethers to form NR formation, then disrupting the integrity of the nuclear envelope structures. would likely increase the NR phenotype, as the structural integrity of the lamin filaments would be less able to counterbalance these NR originates from the nuclear envelope internal mechanical forces. To test this, we expressed the human The mechanism of NR formation is currently unknown. We postulate disease causing lamin protein, progerin by itself and in combination that the NR is a remodeling of the nuclear envelope and therefore with Cap-H2 overexpression. Control cells expressing progerin alone originates from the nuclear envelope. However, it is possible that these exhibited 1.0 NR events per nucleus (Figure 8). Likewise, Cap-H2 structures assemble de novo and are assembled inside the nucleus. To overexpression alone exhibited 1.14 NR events per nucleus (Figure 8). directly address this question, we used GFP-tagged nuclear pore protein However, coexpression of both proteins increased the frequency to Nup107 to observe NR events in live salivary gland cells. After heat 4.49 NR events per nucleus (Figure 8). Based on the results, we have shock induction of Cap-H2, glands were time lapse imaged for 45 min. generated a model illustrating how these various forces may be acting Z-slices were taken to capture three-dimentional structures of the nu- to remodel the nuclear architecture (Figure 9). cleus. During the imaging session, changes in nuclear shape accompa- nied thickening of the nuclear membrane and eventual internalized DISCUSSION budding of the envelope (Figure 7, Figure S6,andFile S1). Control time Nuclear architecture has significant influence over the function of lapses failed to reveal the formation of similar structures (Figure S7 and the nucleus. Beyond its role as an organelle boundary, the nuclear File S2). Interestingly, some NR invaginations already had formed when envelope plays a dynamic role in gene regulation and genomic imaging was initiated and then resorbed back into the envelope, sup- organization. The localization of a gene to the nuclear periphery can porting the idea that NR events are reversible and dynamic (Figure S8 have profound impacts on its transcriptional state or mRNA export and File S3). It is important to note that because structures were imaged efficiency. Therefore, the complex network of nuclear envelope that with z-slices, and that the nucleus remains dynamic during imaging, constitutes the NR may have functional significance, as it brings some features move between individual z-slices. chromatin in the nuclear interior closer in proximity to the inner nuclear envelope. Moreover, defects and instability of the nuclear Perturbing the structural integrity of the nuclear architecture have been directly linked to multiple human diseases, and envelope enhances NR events aberrant regulation of the NR has been observed in several disease Previous studies have suggested that nuclear shape and size are models (Malhas et al. 2011). Thus, understanding the dynamic regu- maintained by a coordinated balance of forces derived from chromatin lation of the NR is relevant to human disease. condensation inside the nucleus and cytoskeletal dynamics out- It has been proposed that forces shape the nuclear envelope and side the nucleus, both converging on the nuclear envelope (Buster changes in these forces can lead to nuclear defects. These forces,

Figure 7 Time-lapse imaging of nucleoplasmic reticulum (NR) for- mation in Cap-H2 overexpressing salivary gland nucleus. Time lapse imaging of the nuclear envelope in Cap-H2 overexpressing nucleus used a fluorescent nuclear en- velope, marked with a green fluo- rescent protein (GFP) tagged nuclear pore complex. Images are displayed in three-minute incre- ments. The nucleus was imaged by capturing z-optical slices, with astepsizeof2microns,images shown are of a single z-slice. Structural changes of the nu- clear envelope can be observed throughout the course of the experiment, including the bud- ding of NR from the nuclear envelope at t = 45’. Arrowhead indicates the location of the NR budding event at initial and final time points. Scale bar is 10 microns. See File S1.

Volume 5 March 2015 | Chromatin Force Remodels the Nuclear Envelope | 349 Our findings indicate that NR formation in Drosophila can occur in unperturbed cells, albeit rarely, and can be mediated by condensin II activity, with significant increase in NR events following induction of Cap-H2. Additionally, we have shown that condensin II-mediated NR events are type II invaginations, indicating that the double mem- brane layer of the nuclear envelope is reshaped to protrude into the nucleoplasm. Live cell analysis, combined with the temporal control of our heat shock expression lines, revealed that the NR is a distortion of the nuclear envelope. These results suggest that NR formation is a dynamic process and that the nuclear envelope is plastic, as it is not restricted to the shape it takes on when the envelope reforms. Previous studies have been unable to temporally resolve NR events. Although it would be difficult to generalize these findings to all NR formation, it is nonetheless pertinent that nuclear architecture can change in response to chromatin state and, in particular, that it is responsive to mechanical forces emanating from within the nucleus. Given the many ways in which chromatin and the nuclear envelope physically interact, we postulated that physical linkages may be responsible for the previously observed NR structures. Experiments that tethered chromatin to the nuclear envelope pre- dictably localized DNA to the nuclear periphery. However, upon heat shock induction of Cap-H2, NR formation was frequently associated with tether attachments. These findings implicate similar endogenous chromatin-envelope interactions in NR formation and are consistent with our electron micrographs that reveal endogenous chromatin-NR contacts. We speculate that these contacts are focal points of NR biogenesis. A recent study has computationally predicted that Dro- sophila polytene chromosomes have approximately 48 chromatin- envelope contacts, many more than previously thought (Kinney et al. 2014). This suggests therefore that many regions of the genome Figure 8 Weakening of the nuclear envelope enhances Cap-H2 induction could function as tethers that drive NR biogenesis. of nucleoplasmic reticulum. Nuclear defects of individual salivary glands Finally, structural weakening of the envelope is able to enhance the overexpressing the human progerin protein are visualized with DAPI and NR phenotype because progerin expression increases NR events. Our anti-Lamin. Overexpression of Cap-H2 yields an enhanced nucleoplasmic fi reticulum (NR) phenotype (A), compared to GAL4 control (B). Quantification ndings support a model whereby nuclear architecture is maintained of NR events in progerin background can increase the phenotype when by a delicate balancing of counteracting forces. By disrupting the paired with Cap-H2 overexpression (C). A significant increase is observed for nuclear lamina, the nuclear envelope becomes more susceptible to overexpression of Cap-H2 and progerin compared to Cap-H2 overexpres- mechanical forces, such as those derived from chromatin reorganiza- sion alone, P-value 4.15e27. Scale bars are 10 microns for all panels. tion. Specifically, we propose that chromatin-envelope tethers trans- duce pulling forces from chromatin condensation that reposition originating from the cytoskeleton and balancing properties of the lamina envelope into the nuclear interior and restructure nuclear architecture. matrix, collectively contribute to erecting the higher order structure of This study and work from others have begun to elucidate the the nuclear envelope. Here, we propose that an additional force acts complexity of factors acting on the nuclear envelope (Dahl et al. upon the nuclear envelope: the mechanical stress exerted from chromo- 2005; Goulbourne et al. 2011; Mazumder and Shivashankar 2007; some condensation and genomic reorganization in interphase cells. Mazumder et al. 2008). The cumulative data suggest a highly dynamic

Figure 9 Model of chromatin compaction leading to nucleo- plasmic reticulum (NR) formation. (A) Attachments between the chromatin (black and gray line) and nuclear envelope (red line) are established and maintained for normal cellular function. (B) As chromatin compaction short- ens the chromatin, the nuclear envelope is pulled into the nuclear space through these attachments. (C) Formation of the NR occurs as the result of one or more chroma- tin-envelope attachments.

350 | J. Bozler et al. nuclear envelope, responding to biological processes throughout in- Gruenbaum, Y., A. Margalit, R. D. Goldman, D. K. Shumaker, and K. L. terphase. Continuing work on the biophysics of the NR as well as the Wilson, 2005 The nuclear lamina comes of age. Nat. Rev. Mol. Cell Biol. genetic factors that regulate these structures will yield valuable insight 6: 21–31. into the maintenance of the nuclear envelope as well as its function Hartl, T. A., H. F. Smith, and G. Bosco, 2008 Chromosome alignment – and role in normal and disease states. and transvection are antagonized by Condensin II. Science 322: 1384 1387. Hirano, T., R. Kobayashi, and M. Hirano, 1997 , chromosome ACKNOWLEDGMENTS condensation protein complexes containing XCAP-C, XCAP-E and We thank Laurie Wallrath and David Tree for generously providing a homolog of the Drosophila Barren protein. Cell 89: 511– fly strains, the Bloomington Drosophila Stock Center and FlyBase for 521. essential resources. This work was supported by an American Federa- Hochstrasser, M., and J. W. Sedat, 1987 Three-dimensional organization of tion of Aging Research BIG grant and National Institutes of Health Drosophila melanogaster interphase nuclei. I. Tissue-specific aspects of GM069462 to G.B., and National Institutes of Health Graduate polytene nuclear architecture. J. Cell Biol. 104: 1455–1470. Training Grant T32GM008704 to J.B. Kimura, K., and T. Hirano, 1997 ATP-dependent positive supercoiling of DNA by 13S condensin: a biochemical implication for chromosome condensation. Cell 90: 625–634. LITERATURE CITED Kinney, N. A., I. V. Sharakhov, and A. V. Onufriev, 2014 Investigation of Bauer, C. R., T. A. Hartl, and G. Bosco, 2012 Condensin II promotes the the chromosome regions with significant affinity for the nuclear envelope formation of chromosome territories by inducing axial compaction of in fruit fly–a model based approach. PLoS One 9: e91943. polyploid interphase chromosomes. PLoS Genet. 8: e1002873. Lammerding, J., L. G. Fong, J. Y. Ji, K. Reue, C. L. Stewart et al., Beard, G. S., J. M. Bridger, I. R. Kill, and D. R. P. Tree, 2008 Towards 2006 Lamins A and C but not lamin B1 regulate nuclear mechanics. a Drosophila model of Hutchinson-Gilford progeria syndrome. Biochem. J. Biol. Chem. 281: 25768–25780. – Soc. Trans. 36: 1389 1392. Malhas, A., C. Goulbourne, and D. J. Vaux, 2011 The nucleoplasmic re- Bermejo, R., T. Capra, R. Jossen, A. Colosio, C. Frattini et al., 2011 The ticulum: form and function. Trends Cell Biol. 21: 362–373. replication checkpoint protects fork stability by releasing transcribed Mammoto, T., and D. E. Ingber, 2010 Mechanical control of tissue and – genes from nuclear pores. Cell 146: 233 246. organ development. Development 137: 1407–1420. Blobel, G., 1985 : a hypothesis. Proc. Natl. Acad. Sci. USA 82: Maniotis, A. J., C. S. Chen, and D. E. Ingber, 1997 Demonstration of me- – 8527 8529. chanical connections between integrins, cytoskeletal filaments, and nu- Budnik, V, Y Zhong, and C F Wu. 1990 Morphological plasticity of motor cleoplasm that stabilize nuclear structure. Proc. Natl. Acad. Sci. USA 94: axons in Drosophila mutants with altered excitability. J. Neurosci. 10: 849–854. – 3754 3768. Mazumder, A., and G. Shivashankar, 2007 Gold-nanoparticle-assisted laser Buster, D. W., S. G. Daniel, H. Q. Nguyen, S. L. Windler, L. C. Skwarek et al., perturbation of chromatin assembly reveals unusual aspects of nuclear 2013 SCFslimb ubiquitin ligase suppresses condensin II-mediated nu- architecture within living cells. Biophys. J. 93: 2209–2216. – clear reorganization by degrading Cap-H2. J. Cell Biol. 201: 49 63. Mazumder, A., T. Roopa, A. Basu, L. Mahadevan, and G. Shivashankar, Capelson, M., Y. Liang, R. Schulte, W. Mair, U. Wagner et al., 2008 Dynamics of chromatin decondensation reveals the structural 2010 Chromatin-bound nuclear pore components regulate gene ex- integrity of a mechanically prestressed nucleus. Biophys. J. 95: 3028–3035. – pression in higher eukaryotes. Cell 140: 372 383. Mazumder, A., T. Roopa, A. Kumar, K. V. Iyer, N. M. Ramdas et al., Chow, K., R. E. Factor, and K. S. Ullman, 2012 The nuclear envelope 2010 Prestressed nuclear organization in living cells. Methods Cell Biol. – environment and its cancer connections. Nat. Rev. Cancer 12: 196 209. 98: 221–239. Dahl, K. N., A. J. Engler, J. D. Pajerowski, and D. E. Discher, 2005 Power- Rajapakse, I., and M. Groudine, 2011 On emerging nuclear order. J. Cell law rheology of isolated nuclei with deformation mapping of nuclear Biol. 192: 711–721. substructures. Biophys. J. 89: 2855–2864. Rodrıguez-Navarro, S., T. Fischer, M. Luo, O. Antúnez, S. Brettschneider de Vosse, D. W. V., Y. Wan, R. W. Wozniak, and J. D. Aitchison, 2011 Role et al., 2004 Sus1, a functional component of the SAGA histone acetylase of the nuclear envelope in genome organization and gene expression. complex and the nuclear pore-associated mRNA export machinery. Cell Wiley Interdiscip. Rev. Syst. Biol. Med. 3: 147–166. 116: 75–86. Dialynas, G., S. Speese, V. Budnik, P. K. Geyer, and L. L. Wallrath, Rowat, A. C., J. Lammerding, H. Herrmann, and U. Aebi, 2008 Towards an 2010 The role of Drosophila Lamin C in muscle function and gene integrated understanding of the structure and mechanics of the cell expression. Development 137: 3067–3077. nucleus. BioEssays 30: 226–236. Dupuy-Coin, A., P. Moens, and M. Bouteille, 1986 Three-dimensional Scaffidi, P., and T. Misteli, 2006 Lamin A-dependent nuclear defects in analysis of given cell structures: nucleolus, nucleoskeleton and nuclear human aging. Science 312: 1059–1063. inclusions. Methods Achiev. Exp. Pathol. 12: 1. Schneider, R., and R. Grosschedl, 2007 Dynamics and interplay of nuclear Fricker, M., M. Hollinshead, N. White, and D. Vaux, 1997 Interphase nuclei architecture, genome organization, and gene expression. Genes Dev. 21: of many mammalian cell types contain deep, dynamic, tubular 3027–3043. membrane-bound invaginations of the nuclear envelope. J. Cell Biol. 136: Shimi, T., V. Butin-Israeli, and R. D. Goldman, 2012 The functions of the 531–544. nuclear envelope in mediating the molecular crosstalk between the nucleus Funkhouser, C. M., R. Sknepnek, T. Shimi, A. E. Goldman, R. D. Goldman and the cytoplasm. Curr. Opin. Cell Biol. 24: 71–78. et al., 2013 Mechanical model of blebbing in nuclear lamin meshworks. Shumaker, D. K., T. Dechat, A. Kohlmaier, S. A. Adam, M. R. Bozovsky et al., Proc. Natl. Acad. Sci. USA 110: 3248–3253. 2006 Mutant nuclear lamin A leads to progressive alterations of epi- George, C. M., J. Bozler, H. Q. Nguyen, and G. Bosco, 2014 Condensins are genetic control in premature aging. Proc. Natl. Acad. Sci. USA 103: 8703– required for maintenance of nuclear architecture. Cells 3: 865–882. 8708. Goldman, R. D., D. K. Shumaker, M. R. Erdos, M. Eriksson, A. E. Goldman Sullivan, T., D. Escalante-Alcalde, H. Bhatt, M. Anver, N. Bhat et al., et al., 2004 Accumulation of mutant lamin A causes progressive 1999 Loss of A-type lamin expression compromises nuclear envelope changes in nuclear architecture in Hutchinson–Gilford progeria syn- integrity leading to muscular dystrophy. J. Cell Biol. 147: 913–920. drome. Proc. Natl. Acad. Sci. USA 101: 8963–8968. Therizols, P., C. Fairhead, G. G. Cabal, A. Genovesio, J. Olivo-Marin et al., Goulbourne, C. N., A. N. Malhas, and D. J. Vaux, 2011 The induction of 2006 Telomere tethering at the nuclear periphery is essential for effi- a nucleoplasmic reticulum by prelamin A accumulation requires CTP: cient DNA double strand break repair in subtelomeric region. J. Cell Biol. phosphocholine cytidylyltransferase-alpha. J. Cell Sci. 124: 4253–4266. 172: 189–199.

Volume 5 March 2015 | Chromatin Force Remodels the Nuclear Envelope | 351 Verdaasdonk, J., P. Vasquez, R. Barry, T. Barry, S. Goodwin et al., proteins and inner nuclear LBR. J. Biol. Chem. 272: 2013 Centromere tethering confines chromosome domains. Mol. Cell 14983–14989. 52: 819–831. Zhao, K., A. Harel, N. Stuurman, D. Guedalia, and Y. Gruenbaum, Wallace, H. A., and G. Bosco, 2013 Condensins and 3D organization of 1996 Binding of matrix attachment regions to nuclear lamin is medi- theinterphasenucleus.CurrentGeneticMedicineReports1:219– ated by the rod domain and depends on the lamin polymerization state. 229. FEBS Lett. 380: 161–164. Ye, Q., I. Callebaut, A. Pezhman, J. Courvalin, and H. J. Worman, 1997 Domain-specific interactions of human HP1-type chromodomain Communicating editor: J. A. Birchler

352 | J. Bozler et al.